[HOT] West Africa Ebola Activation : follow-up

Pierre Béland pierzenh at yahoo.fr
Wed Sep 3 16:30:51 UTC 2014


how do you process the update of the polygons to remove the place tag in this? A second file that you will produce to edit these?


 De : Rafael Avila Coya <ravilacoya at gmail.com>
À : hot at openstreetmap.org 
Envoyé le : Mercredi 3 septembre 2014 12h04
Objet : Re: [HOT] West Africa Ebola Activation : follow-up

Hash: SHA1

Hi, Andrew:

I can start tomorrow early in the morning (it's now midnight in the
Philippines and I am going to sleep).

Could you produce that osm file with the residential polygons (only
those that have names of course)? Better if you split them in three
files, one for each country, so it's easier to process them. I am in a
tiny island using 3G, so I could run out of data easily. The files,
you can upload them to dropbox or something similar. With luck, I
could maybe have the result file during the morning, but never promise

About the pgp, I guess I have to update my system...



On 03/09/14 22:30, Andrew Buck wrote:
> Yeah, this sounds like a good process to me.  For the area for the 
> overpass query it would be Guinea, Liberia, and Sierra Leone.  We
> have used the 'area' query in the past with the relation id of the
> country in question with good results using this method.  The
> process sounds pretty good.  I would say that almost all of the
> areas would be either hamlet or village at most so we would need to
> tweak that slightly, or even just make them all hamlet (which is
> what we are doing with the GNS nodes we are merging in).  All of
> these will likely need to be classified anyway so it is hard to
> decide how to classify them at this point when the others haven't
> already been done either.
> On a different topic, I noticed the pgp signature on your last
> email did not verify correctly (others have in the past), nothing
> to worry about I think, just thought I would mention it.  A bit
> odd.
> -AndrewBuck
> On 09/03/2014 09:19 AM, Rafael Avila Coya wrote:
>> Hi Andrew and Pierre:
>> The idea would be more or less like this:
>> Download the 4,500 target polygons with overpass or similar.
>> Maybe one of you who know the area where they are can do it.
>> Then I would generate an osm file with the script (it works with 
>> ways as well as with relations). This osm file would have only
>> the nodes for the place, with the name tag and the kind of place,
>> based on the area. At this moment, the script decides only
>> between village or town, depending on the area (because that was
>> the only need for that specific Nigeria import), but I guess it
>> wouldn't be difficult to tweak the code to produce other kind of
>> places, like hamlet or city. Alternatively, we could just set the
>> place tag as place=fixme or something, and let the contributors
>> decide which is the most suitable node by node, although that
>> would slow the process.
>> The job could be set to be done with the TM, similarly as with
>> the UNICEF import in CAR, or the eHealth Africa Health
>> Facilities imports in Nigeria. Example:
>> http://tasks.hotosm.org/project/637 A user takes a task and
>> checks node by node that they are ok, and reports any problem in
>> the task comment (if a node was already in OSM, it ignores it
>> (deletes it)). I've set already a few jobs of this kind (the last
>> one a few days ago only), so I can do that myself if you want.
>> Finally, once all the job is finished, we can download the 
>> polygons again with overpass or mirror download, delete all the 
>> names for the residential polygons at once using filters and
>> upload the changes.
>> I maybe missing something here, anyway...
>> Cheers,
>> Rafael.
>> On 03/09/14 21:24, Pierre Béland wrote:
>>> Hi Rafael,
>>> this script look interesting, taking care to remove the tag and
>>>  place it on a node.
>>> Since there might be names around to reconcile with, this
>>> import cannot be done mecanically.
>>> There are some 4,500 of these places on polygons. To share this
>>>  task among many contributors, I think that a good solution
>>> would be that the output of your script split the list of
>>> polygons + places, with a maximum  of 100 polygons / places per
>>> file,
>>> This way, the JOSM Todo plugin would let go rapidly through the
>>>  list of place nodes, merge with an other place if appropriate
>>> and upload the result to the OSM database.
>>> Is this make sense?
>>> Pierre
>>> ------------------------------------------------------------------------
>> *De :* Rafael Avila Coya <ravilacoya at gmail.com>
>>> *À :* hot at openstreetmap.org *Envoyé le :* Mercredi 3 septembre
>>>  2014 2h25 *Objet :* Re: [HOT] West Africa Ebola Activation : 
>>> follow-up
>>> Hi, Pierre:
>>> I wonder if you thought about replacing the names on the 
>>> residential polygons by a node in their centroid automatically
>>>  using a script.
>>> I created one for a similar task: the import of place names for
>>>  several states in North Nigeria, still a work in progress
>>> [1]. For that, I wrote a script that extracts the names of the
>>> places and creates an osm file with the nodes.
>>> If needed, I can tweak the code to make one for this ebola
>>> task.
>>> Cheers,
>>> Rafael.
>>> [1] 
>>> https://wiki.openstreetmap.org/wiki/Import_Nigeria_eHealth_Africa_Places
>> On 03/09/14 06:45, Pierre Béland wrote:
>>>> From march to sept. 1,  there are 1,201 OSM contributors who
>>>>  edited 6.7 million objects for Guinea, Liberia and Sierra 
>>>> Leone. But reading the news, you understand that we cannot
>>>> stop there. Humanitarian organizations are operating in a
>>>> very difficult context and count on our support. We receive 
>>>> constantly messages of gratitude, people saying how useful
>>>> is this information for them plus all the export services 
>>>> provided.
>>>> Plus a vast plan of support by the various UN Agencies is 
>>>> presently organized. Andrew Buck and I ha today a meeting
>>>> with GIS staff from World Health Organization (WHO), Office
>>>> for the Coordination of Humanitarian Affairs (OCHA) and the
>>>> World food program (WFP). After looking at the various
>>>> possibilities, it has been proposed to use OpenStreetMap for
>>>> the base map of the various agencies.
>>>> In this context, it will be important to revise / complete
>>>> both administrative boundaries and populated places.
>>>> To plan their intervention, the humanitarian organizations
>>>> need administrative boundaries at various levels of
>>>> administration and the Populated place names. We have already
>>>> revised the  the counties and added district boundaries for
>>>> the Liberia and are looking at the other countries.
>>>> If any of you have contacts in Sierra Leone or Guinea to try
>>>>  obtain administrative boundary files with the appropriate 
>>>> license for OSM, this would greatly help.
>>>> For the populated places, we agreed to move any place name 
>>>> which is on a residential polygon to a node. This will
>>>> simplify the synchronization with the databases of our
>>>> various partners. We are also looking at adding a pcode like
>>>> we did for Haii in 2010. There are 4,500 place tags on
>>>> residential polygons to be converted to nodes.
>>>> Plus there are two Task Manager jobs still to be completed to
>>>>  import place names for Liberia and Sierra Leone. These two 
>>>> tasks needs some care and we are asking again for
>>>> contributors familiar enough with JOSM to contact either me
>>>> or Andrew. Since these are private jobs restricted to a few
>>>> persons, you need to provide us you OpenStreetMap account. 
>>>> http://tasks.hotosm.org/project/617
>>> <http://tasks.hotosm.org/project/617>
>>>> http://tasks.hotosm.org/project/619
>>>> Blake Girardot can also help people to learn this process.
>>>> For those that prefer to do general mapping, digitizing from 
>>>> Bing Imagery, look at the task for Eastern Liberia counties.
>>>> As the epidemy is still progressing, we will also add task
>>>> for the north-east Nimba county.
>>>> http://tasks.hotosm.org/project/624
>>>> Pierre
>>>> _______________________________________________ HOT mailing 
>>>> list HOT at openstreetmap.org <mailto:HOT at openstreetmap.org>
>>> https://lists.openstreetmap.org/listinfo/hot
>>> _______________________________________________ HOT mailing
>>> list HOT at openstreetmap.org <mailto:HOT at openstreetmap.org> 
>>> https://lists.openstreetmap.org/listinfo/hot
>> -- Twitter: http://twitter.com/ravilacoya
>> --------------------------------
>> Por favor, non me envíe documentos con extensións .doc, .docx, 
>> .xls, .xlsx, .ppt, .pptx, aínda podendoo facer,  non os abro.
>> Atendendo á lexislación vixente, empregue formatos estándares e 
>> abertos.
>> http://es.wikipedia.org/wiki/OpenDocument#Tipos_de_ficheros
>> _______________________________________________ HOT mailing list
>>  HOT at openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/hot
> _______________________________________________ HOT mailing list 
> HOT at openstreetmap.org https://lists.openstreetmap.org/listinfo/hot

- -- 
Twitter: http://twitter.com/ravilacoya

- --------------------------------

Por favor, non me envíe documentos con extensións .doc, .docx, .xls,
.xlsx, .ppt, .pptx, aínda podendoo facer,  non os abro.

Atendendo á lexislación vixente, empregue formatos estándares e abertos.

Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with undefined - http://www.enigmail.net/



HOT mailing list
HOT at openstreetmap.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/hot/attachments/20140903/3bbb6f57/attachment-0001.html>

More information about the HOT mailing list